Applications monitor many metrics of a computing system. Metrics include, for example, a response time of a component of a computing system, a throughput of a component, and an error rate of a component. As computing systems grow larger and move to cloud environments, the number of metrics that may be monitored and the volume of data that may be detected is increasing.
The approaches described in this section are approaches that could be pursued, but not necessarily approaches that have been previously conceived or pursued. Therefore, unless otherwise indicated, it should not be assumed that any of the approaches described in this section qualify as prior art merely by virtue of their inclusion in this section.